Instructions (SCMG Week 3) Week 3: Judgmental Forecasting Course Objectives: CO3: Evaluate Forecasting methods, predictability factors, and causes of error in Forecasting Prompt: This week you will
Judgmental Forecasting Course and Assignment Student Name Overview & Introduction Use this section to introduce your paper and the resources you will use to describe Judgmental Forecasting. You may erase this notation, but only up to where the sentence starts or you will eliminate the formatting. What is Judgmental Forecasting The heading above would be used to divide your paper into sections based on content. This is the first level of heading, and it is centered and bolded with each word of four letters or more capitalized. The heading should be a short descriptor of the section. Most papers will have headings or subheadings to organize the paper into sections. Where is Judgmental Forecasting used Most Effectively? The subheading above would be used if there are several sections within the topic labeled in a heading. The subheading is flush left and bolded, with each word of four letters or more capitalized. Examples of how Judgmental Forecasting is Used Most Effectively Xxx Advantages and Disadvantages of Judgmental Forecasting Xxx Advantages Xxx Disadvantages Xxx The Delphi Method of Judgmental Forecasting In this section, thoroughly explain the use of, advantages of, and an example of the Delphi method of forecasting. You may erase these instructions prior to inserting your narrative. References (Example of a web-based Reference) U.S. Department of Health and Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute. (2003). Managing asthma: A guide for schools (NIH Publication No. 02-2650). Retrieved from http://www.nhlbi.nih.gov/health/prof/asthma/asth_sch.pdf This is a correctly cited Web address
Instructions (SCMG Week 3) Week 3: Judgmental Forecasting Course Objectives: CO3: Evaluate Forecasting methods, predictability factors, and causes of error in Forecasting Prompt: This week you will
Ch ap te r 4 J u d gm en ta l f o re ca sts Fore ca stin g u sin g j u dgem en t i s c o m mon i n p ra ctic e . I n m an y c a se s, j u dgm en ta l fo re ca stin g i s t h e o n ly o p tio n , s u ch a s w hen t h ere i s a c o m ple te l a ck o f h is to ric a l d ata , o r w hen a n ew p ro d u ct i s b ein g l a u n ch ed , o r w hen a n ew c o m petit o r e n te rs t h e m ark et, o r d u rin g c o m ple te ly n ew a n d u n iq u e m ark et c o n dit io n s. F or e x am ple , i n D ece m ber 2 0 12 , t h e A ustr a lia n g o vern m en t w as t h e f ir s t i n t h e w orld t o p ass l e g is la tio n t h at b an n ed t h e u se o f c o m pan y l o go s o n c ig are tte p ack ets , a n d r e q u ir e d a ll c ig are tte p ack ets t o b e a d ark g re en c o lo u r. J u dgem en t m ust b e a p plie d i n o rd er t o f o re ca st t h e effe ct o f s u ch a p olic y , a s t h ere a re n o h is to ric a l p re ce d en ts . T here a re a ls o s it u atio n s w here t h e d ata a re i n co m ple te , o r o n ly b eco m e a v aila b le a ft e r so m e d ela y . F or e x am ple , c e n tr a l b an ks i n clu de j u dgem en t w hen f o re ca stin g t h e cu rr e n t l e v el o f e co n om ic a ctiv it y , a p ro ce d u re k n ow n a s n ow ca stin g, a s G D P i s o n ly a v aila b le o n a q u arte rly b asis . R ese a rc h i n t h is a re a h as s h ow n t h at t h e a ccu ra cy o f j u dgm en ta l f o re ca stin g i m pro ves w hen t h e f o re ca ste r h as ( i) i m porta n t d om ain k n ow le d ge, a n d ( ii) m ore t im ely , u p-to – d ate i n fo rm atio n . A j u dgm en ta l a p pro ach c a n b e q u ic k t o a d ju st t o s u ch c h an ges, in fo rm atio n o r e v en ts . O ver t h e y ea rs , t h e a cce p ta n ce o f j u dgm en ta l f o re ca stin g a s a s c ie n ce h as i n cre a se d , a s h as t h e r e co gn it io n o f i t s n eed . M ore i m porta n tly , t h e q u alit y o f j u dgm en ta l f o re ca sts h as a ls o i m pro ved , a s a d ir e ct r e su lt o f r e co gn is in g t h at i m pro vem en ts i n j u dgm en ta l fo re ca stin g c a n b e a ch ie v ed b y i m ple m en tin g w ell- str u ctu re d a n d s y ste m atic a p pro ach es. I t i s i m porta n t t o r e co gn is e t h at j u dgm en ta l f o re ca stin g i s s u bje ctiv e a n d co m es w it h l im it a tio n s. H ow ev er, i m ple m en tin g s y ste m atic a n d w ell- str u ctu re d a p pro ach es c a n c o n fin e t h ese l im it a tio n s a n d m ark ed ly i m pro ve f o re ca st a ccu ra cy . T here a re t h re e g en era l s e ttin gs i n w hic h j u dgm en ta l f o re ca stin g i s u se d : ( i) t h ere a re n o a v aila b le d ata , s o t h at s ta tis tic a l m eth od s a re n ot a p plic a b le a n d j u dgm en ta l fo re ca stin g i s t h e o n ly f e a sib le a p pro ach ; ( ii) d ata a re a v aila b le , s ta tis tic a l f o re ca sts a re g en era te d , a n d t h ese a re t h en a d ju ste d u sin g j u dgem en t; a n d ( ii i) d ata a re a v aila b le a n d sta tis tic a l a n d j u dgm en ta l f o re ca sts a re g en era te d i n dep en den tly a n d t h en c o m bin ed . 4 We s h ou ld c la rif y t h at w hen d ata a re a v aila b le , a p ply in g s ta tis tic a l m eth od s ( s u ch a s th ose d is c u sse d i n o th er c h ap te rs o f t h is b o ok), i s p re fe ra b le a n d s h ou ld a lw ay s b e u se d a s a s ta rtin g p oin t. S ta tis tic a l f o re ca sts a re g en era lly s u perio r t o g en era tin g f o re ca sts u sin g o n ly j u dgem en t. F or t h e m ajo rit y o f t h e c h ap te r, w e f o cu s o n t h e f ir s t s e ttin g w here n o d ata a re a v aila b le , a n d i n t h e l a st s e ctio n w e d is c u ss t h e j u dgm en ta l ad ju stm en t o f s ta tis tic a l f o re ca sts . W e d is c u ss c o m bin in g f o re ca sts i n S ectio n 12 .4 . B ib lio gra p hy Law re n ce , M ., G ood w in , P ., O ’C on nor, M ., & Ö nkal, D . ( 2 0 06 ). J u dgm en ta l f o re ca stin g: A r e v ie w o f p ro gre ss o ver t h e l a st 2 5 y ea rs . I n te rn atio n al J o u rn al o f F ore ca stin g , 2 2 (3 ), 4 9 3– 518 . [ D OI] 4 . L aw re n ce , G ood w in , O ’C on nor, & Ö nkal ( 2 0 06 )
Instructions (SCMG Week 3) Week 3: Judgmental Forecasting Course Objectives: CO3: Evaluate Forecasting methods, predictability factors, and causes of error in Forecasting Prompt: This week you will
4.3 T h e D elp hi m eth od The D elp h i m eth od w as i n ven te d b y O la f H elm er a n d N orm an D alk ey o f t h e R an d Corp ora tio n i n t h e 1 9 50 s f o r t h e p u rp ose o f a d dre ssin g a s p ecif ic m ilit a ry p ro ble m . T he m eth od r e lie s o n t h e k ey a ssu m ptio n t h at f o re ca sts f r o m a g ro u p a re g en era lly m ore a ccu ra te t h an t h ose f r o m i n div id u als . T he a im o f t h e D elp h i m eth od i s t o c o n str u ct co n se n su s f o re ca sts f r o m a g ro u p o f e x p erts i n a s tr u ctu re d i t e ra tiv e m an ner. A f a cilit a to r i s a p poin te d i n o rd er t o i m ple m en t a n d m an ag e t h e p ro ce ss. T he D elp h i m eth od g en era ll y i n vo lv es t h e f o llo w in g s ta g es: 1. A p an el o f e x p erts i s a sse m ble d . 2. F ore ca stin g t a sk s/c h alle n ges a re s e t a n d d is tr ib u te d t o t h e e x p erts . 3 . E xp erts r e tu rn i n it ia l f o re ca sts a n d j u stif ic a tio n s. T hese a re c o m pile d a n d su m m aris e d i n o rd er t o p ro vid e f e ed back . 4 . F eed back i s p ro vid ed t o t h e e x p erts , w ho n ow r e v ie w t h eir f o re ca sts i n l ig h t o f t h e fe ed back . T his s te p m ay b e i t e ra te d u n til a s a tis fa cto ry l e v el o f c o n se n su s i s r e a ch ed . 5. F in al f o re ca sts a re c o n str u cte d b y a g gre g atin g t h e e x p erts ’ f o re ca sts . E ach s ta g e o f t h e D elp h i m eth od c o m es w it h i t s o w n c h alle n ges. I n w hat f o llo w s, w e pro vid e s o m e s u ggestio n s a n d d is c u ssio n s a b o u t e a ch o n e o f t h ese . E xp erts a n d a n onym it y T he f ir s t c h alle n ge o f t h e f a cilit a to r i s t o i d en tif y a g ro u p o f e x p erts w ho c a n c o n tr ib u te t o t h e f o re ca stin g t a sk . T he u su al s u ggestio n i s s o m ew here b etw een 5 a n d 2 0 e x p erts w it h d iv ers e e x p ertis e . E xp erts s u bm it f o re ca sts a n d a ls o p ro vid e d eta ile d q u alit a tiv e ju stif ic a tio n s f o r t h ese . A k ey f e a tu re o f t h e D elp h i m eth od i s t h at t h e p artic ip atin g e x p erts r e m ain a n on ym ou s at a ll t im es. T his m ea n s t h at t h e e x p erts c a n not b e i n flu en ce d b y p olit ic a l a n d s o cia l p re ssu re s i n t h eir f o re ca sts . F u rth erm ore , a ll e x p erts a re g iv en a n e q u al s a y a n d a ll a re h eld a cco u n ta b le f o r t h eir f o re ca sts . T his a v o id s t h e s it u atio n w here a g ro u p m eetin g i s h eld a n d s o m e m em bers d o n ot c o n tr ib u te , w hile o th ers d om in ate . I t a ls o p re v en ts 7 mem bers e x ertin g u n du e i n flu en ce b ase d o n s e n io rit y o r p ers o n alit y . T here h av e b een su gg estio n s t h at e v en s o m eth in g a s s im ple a s t h e s e a tin g a rra n gem en ts i n a g ro u p se tt in g c a n i n flu en ce t h e g ro u p d yn am ic s. F u rth erm ore , t h ere i s a m ple e v id en ce t h at a g ro u p m eetin g s e ttin g p ro m ote s e n th usia sm a n d i n flu en ce s i n div id u al j u dgem en t, le a d in g t o o p tim is m a n d o verc o n fid en ce . A b y- pro d u ct o f a n on ym it y i s t h at t h e e x p erts d o n ot n eed t o m eet a s a g ro u p i n a p h ysic a l l o ca tio n . A n i m porta n t a d van ta g e o f t h is i s t h at i t i n cre a se s t h e l ik elih ood o f gath erin g e x p erts w it h d iv ers e s k ills a n d e x p ertis e f r o m v ary in g l o ca tio n s. F u rth erm ore , i t m ak es t h e p ro ce ss c o st- effe ctiv e b y e lim in atin g t h e e x p en se a n d in co n ven ie n ce o f t r a v el, a n d i t m ak es i t f le x ib le , a s t h e e x p erts o n ly h av e t o m eet a c o m m on d ea d lin e f o r s u bm it tin g f o re ca sts , r a th er t h an h av in g t o s e t a c o m mon m eetin g t im e. S ett in g t h e f o re ca stin g t a sk i n a D elp hi In a D elp h i s e ttin g, i t m ay b e u se fu l t o c o n du ct a p re lim in ary r o u n d o f i n fo rm atio n gath erin g f r o m t h e e x p erts b efo re s e ttin g t h e f o re ca stin g t a sk s. A lt e rn ativ ely , a s ex p erts s u bm it t h eir i n it ia l f o re ca sts a n d j u stif ic a tio n s, v alu ab le i n fo rm atio n w hic h i s n ot s h are d b etw een a ll e x p erts c a n b e i d en tif ie d b y t h e f a cilit a to r w hen c o m pilin g t h e fe ed back . F e ed back Feed back t o t h e e x p erts s h ou ld i n clu de s u m mary s ta tis tic s o f t h e f o re ca sts a n d o u tlin es o f q u alit a tiv e j u stif ic a tio n s. N um eric a l d ata s u m marie s a n d g ra p h ic a l r e p re se n ta tio n s ca n b e u se d t o s u m maris e t h e e x p erts ’ f o re ca sts . A s t h e f e ed back i s c o n tr o lle d b y t h e f a cilit a to r, t h ere m ay b e s c o p e t o d ir e ct a tte n tio n an d i n fo rm atio n f r o m t h e e x p erts t o a re a s w here i t i s m ost r e q u ir e d . F or e x am ple , t h e fa cilit a to r m ay d ir e ct t h e e x p erts ’ a tte n tio n t o r e sp on se s t h at f a ll o u ts id e t h e in te rq u artile r a n ge, a n d t h e q u alit a tiv e j u stif ic a tio n f o r s u ch f o re ca sts . 8 It e ra tio n The p ro ce ss o f t h e e x p erts s u bm it tin g f o re ca sts , r e ce iv in g f e ed back , a n d r e v ie w in g th eir f o re ca sts i n l ig h t o f t h e f e ed back , i s r e p ea te d u n til a s a tis fa cto ry l e v el o f co n se n su s b etw een t h e e x p erts i s r e a ch ed . S atis fa cto ry c o n se n su s d oes n ot m ea n co m ple te c o n verg en ce i n t h e f o re ca st v alu e; i t s im ply m ea n s t h at t h e v aria b ilit y o f t h e re sp on se s h as d ecre a se d t o a s a tis fa cto ry l e v el. U su ally t w o o r t h re e r o u n ds a re s u ff ic ie n t. E xp erts a re m ore l ik ely t o d ro p o u t a s t h e n um ber o f i t e ra tio n s i n cre a se s, s o to o m an y r o u n ds s h ou ld b e a v o id ed . F in al f o re ca sts T he f in al f o re ca sts a re u su ally c o n str u cte d b y g iv in g e q u al w eig h t t o a ll o f t h e e x p erts ’ fo re ca sts . H ow ev er, t h e f a cilit a to r s h ou ld k eep i n m in d t h e p ossib ilit y o f e x tr e m e v alu es w hic h c a n d is to rt t h e f in al f o re ca st. L im it a tio n s a n d v a ria tio n s Apply in g t h e D elp h i m eth od c a n b e t im e c o n su m in g. I n a g ro u p m eetin g, f in al f o re ca sts c a n p ossib ly b e r e a ch ed i n h ou rs o r e v en m in ute s — s o m eth in g w hic h i s a lm ost im possib le t o d o i n a D elp h i s e tt in g. I f i t i s t a k in g a l o n g t im e t o r e a ch a c o n se n su s i n a D elp h i s e ttin g, t h e p an el m ay l o se i n te re st a n d c o h esiv en ess. In a g ro u p s e ttin g, p ers o n al i n te ra ctio n s c a n l e a d t o q u ic k er a n d b ette r c la rif ic a tio n s o f q u alit a tiv e j u stif ic a tio n s. A v aria tio n o f t h e D elp h i m eth od w hic h i s o ft e n a p plie d i s t h e “estim ate – ta lk – estim ate ” m eth od , w here t h e e x p erts c a n i n te ra ct b etw een i t e ra tio n s, a lt h ou gh t h e f o re ca st s u bm is sio n s c a n s till r e m ain a n on ym ou s. A d is a d van ta g e o f t h is v aria tio n i s t h e p ossib ilit y o f t h e l o u dest p ers o n e x ertin g u n du e i n flu en ce . T h e f a cil it a to r The r o le o f t h e f a cilit a to r i s o f t h e u tm ost i m porta n ce . T he f a cilit a to r i s l a rg ely r e sp on sib le f o r t h e d esig n a n d a d m in is tr a tio n o f t h e D elp h i p ro ce ss. T he f a cilit a to r i s a ls o r e sp on sib le f o r p ro vid in g f e ed back t o t h e e x p erts a n d g en era tin g t h e f in al fo re ca sts . I n t h is r o le , t h e f a cilit a to r n eed s t o b e e x p erie n ce d e n ou gh t o r e co gn is e a re a s th at m ay n eed m ore a tte n tio n , a n d t o d ir e ct t h e e x p erts ’ a tte n tio n t o t h ese . A ls o , a s th ere i s n o f a ce – to – fa ce i n te ra ctio n b etw een t h e e x p erts , t h e f a cilit a to r i s r e sp on sib le f o r d is se m in atin g i m porta n t i n fo rm atio n . T he e ffic ie n cy a n d e ffe ctiv en ess o f t h e fa cilit a to r c a n d ra m atic a lly i n cre a se t h e p ro bab ilit y o f a s u cce ssfu l D elp h i m eth od i n a j u dgm en ta l f o re ca stin g s e ttin g. B ib lio gra p hy Bueh le r, R ., M esse rv ey , D ., & G rif fin , D . ( 2 0 05). C olla b o ra tiv e p la n nin g a n d p re d ic tio n : D oes g ro u p d is c u ssio n a ffe ct o p tim is tic b ia se s i n t im e e stim atio n ? O rg an iz a tio n al B eh av io r a n d H um an D ecis io n P ro ce sse s , 9 7 (1 ), 4 7– 63. [ D OI] R ow e, G . ( 2 0 07). A g u id e t o D elp h i . F ore sig h t: T he I n te rn atio n al J o u rn al o f A pplie d F ore ca stin g , ( 8 ), 1 1– 16 . R ow e, G ., & W rig h t, G . ( 1 9 99). T he D elp h i t e ch n iq u e a s a f o re ca stin g t o ol: I s su es a n d an aly sis . I n te rn atio n al J o u rn al o f F ore ca stin g , 1 5 (4 ), 3 5 3– 37 5 . [ D OI] 7 . F or f u rth er r e a d in g, r e fe r t o : R ow e ( 2 0 07 ); R ow e & W rig h t ( 1 9 99 ) 8. B ueh le r, M esse rv ey , & G rif fin ( 2 0 05 )